Free Auto Caption Generator with AI
Transcribe speech, style captions and export MP4 or SRT. No upload, no watermark, no sign-up.
Generate automatic captions for any video in your browser. AI speech recognition writes word-timed captions you can edit, restyle and export as a burned-in MP4 or SRT and VTT subtitle files. Your video is never uploaded. No sign-up, no watermark.
- No upload
- No watermark
- No sign-up
- Word-level timing
- SRT and VTT
- 40+ styles
How do I generate captions for a video automatically and for free?
Drop your video into the tool above, press Generate Captions and let the on-device AI transcribe the speech. Review and fix any words, choose a caption style, then export a burned-in MP4 or download SRT and VTT subtitle files. Everything runs in your browser, so nothing is uploaded and exports have no watermark.
- Price
- Free, no credits and no watermark
- Account required
- No
- Video uploaded to a server
- No, transcription runs in your browser
- Input
- MP4, WebM, MOV, MKV (if your browser can play it)
- Output
- Burned-in MP4; SRT and VTT subtitle files
- AI caption language
- English speech today. Subtitle files in any language can be imported in the Video Editor.
- Caption timing
- Word-level timing
- Caption styles
- 40+ styles, fully customisable font, colour, outline, highlight and position
- File size limit
- 1 GB per imported file
- Best browsers
- Chrome and Edge on desktop; Firefox works with the compatibility engine
- Offline
- After the model is downloaded, transcription and export run locally
- Not included
- Cloud templates, stock assets, multilingual AI captions, motion tracking
What is the ToolDeck auto caption generator?
ToolDeck Auto Caption Generator is a free, browser-based tool that turns speech into timed captions using open-source Whisper speech recognition. It runs on your own device, so your video is never uploaded. You can edit every caption line, apply 40+ styles, and export a burned-in MP4 or download the captions as SRT and VTT subtitle files. It needs no account and adds no watermark.
It is built for short-form social video, tutorials, talking-head clips and private footage. It recognises English speech today. For subtitles in other languages, use the Video Editor to import an SRT or VTT file. It is not a cloud editor and does not include stock assets, templates or motion tracking.
How to auto-generate captions in 3 steps
- 1
Drop in your video
Open the tool and drop an MP4, WebM or MOV file. The file is read in your browser and is not uploaded.
- 2
Generate AI captions
Press Generate Captions. Whisper transcribes the speech on your device, times every word and adds editable caption clips. The first run downloads the model once; later runs are faster.
- 3
Fix, style and export
Review the text, fix any words, choose a caption style and position, then export a burned-in MP4 or download SRT and VTT subtitle files.
Caption features built for real videos
AI auto captions that run on your device
Open-source Whisper speech recognition transcribes English speech into timed caption clips. The model downloads once and is cached in your browser. Your audio never leaves your device.
Word-level timing and editable captions
Every word is timed. Click any caption line to fix a word, adjust the timing, split a line or delete it. The preview updates as you type, so you see exactly what exports.
Burn captions into MP4 or export SRT and VTT
Download a burned-in MP4 with captions drawn on the picture, or download the same captions as SRT or VTT subtitle files with identical timings for YouTube, Premiere, DaVinci and CapCut.
40+ caption styles and full customisation
Pick from bold, clean, animated and word-by-word styles. Change font, size, colour, outline, highlight, glow and position. Drag captions anywhere on the video.
Short-form and long-form caption modes
Choose Short captions for 9:16 TikTok, Reels and Shorts, or Standard for YouTube and tutorials. You can switch modes after generating captions without re-transcribing.
No upload, no sign-up, no watermark
Everything runs inside your browser tab. There is no account, no email wall, no credit system and no watermark on exports. The tool is free and private by design.
What you can caption with this tool
TikTok, Reels and Shorts
Generate word-by-word captions in a bold style, burn them into a 9:16 MP4 and post. No watermark, no app install.
YouTube videos and tutorials
Generate captions on your device, fix names and technical terms, then download an SRT for YouTube or a burned-in MP4.
Accessibility and silent viewing
Most social video is watched with the sound off. Captions keep viewers engaged and make your video accessible.
Client and private footage
Caption interviews, internal training or family videos without sending a single file to a server.
Podcast clips and talking-head videos
Turn speech into readable captions quickly, then export the subtitle file for your editor or upload it with the video.
Multilingual subtitle workflows
If your speech is English, generate captions here; for other languages, import an SRT or VTT file in the video editor and burn it in.
Auto caption generator FAQ
Answers about pricing, privacy, caption accuracy, SRT and VTT export, styles and browsers.
Free, watermark and privacy
Is this auto caption generator really free?
Yes. ToolDeck sets no credits, no daily cap and no sign-up. The tool adds no watermark to captioned videos. The practical limits come from your device: browser memory, processor speed and a 1 GB limit per imported file.
Does it upload my video?
No. Transcription and export run inside your browser tab. Your video and audio are never sent to ToolDeck servers. To verify, open your browser developer tools, go to the Network tab, then generate captions and export a video.
Do I need an account or to install an app?
No. Open the page, drop in your video and generate captions. There is no login, no email wall, no trial and no install.
Is it a free alternative to CapCut auto captions?
It covers the core job: automatic captions with word-level timing and SRT/VTT export, with no watermark. CapCut offers multilingual captions and cloud templates that ToolDeck does not. ToolDeck runs locally and needs no account, which makes it a strong fit for private footage and quick caption jobs.
Captions and subtitles
How do I generate captions for a video automatically?
Drop your video into the tool, press Generate Captions and wait for Whisper to transcribe the speech. Captions appear with word-level timing. You can fix words, restyle captions and export a burned-in MP4 or download SRT and VTT files.
Can I edit the captions after generating them?
Yes. Every caption line is editable. Click a line to fix a word, adjust the start and end time, split a line or delete it. The preview updates as you type.
What is the difference between burned-in captions and SRT files?
Burned-in captions are drawn on the video picture and always appear, which is why social video uses them. SRT and VTT files are separate subtitle files that viewers can switch on or off and that editors like Premiere, DaVinci and CapCut can import. You can export both from the same captions.
Which languages does the AI caption generator support?
The built-in speech model recognises English speech today. Accuracy drops with background noise, strong accents and music. For subtitle files in other languages, open the Video Editor and import an SRT or VTT file.
Can I export captions for YouTube, Premiere or CapCut?
Yes. Download the captions as SRT or VTT with identical timings. Upload the SRT to YouTube under Subtitles, or import it into Premiere Pro, DaVinci Resolve, CapCut or any editor that accepts subtitle files.
Caption styles and export
What caption styles are available?
40+ styles, including clean white captions, bold word-by-word highlighting, animated and outline styles. You can change the font, size, colour, outline, highlight, glow and position, and drag captions anywhere on the video.
What export formats are available?
Export a burned-in MP4 with captions drawn on the picture, or download the captions as SRT or VTT subtitle files. The MP4 export uses the same captions and timings you see in the preview.
Does it work on iPhone and Android?
The page loads on mobile, but the layout and performance are built for desktop. For serious captioning and exports, use Chrome or Edge on a computer.
Why is transcription slow?
Whisper runs on your device. The first run downloads the speech model, which takes longer. After that, transcription is faster because the model is cached. Long videos take longer than short clips, and a device without a GPU runs on the CPU.
What this caption generator does well, and where it is limited
Everything runs on your own computer, so results depend on audio quality, your browser and your device.
| Scenario | Result | Notes |
|---|---|---|
| Clear English speech | Works well | Accuracy is good for clear audio. Review names and technical terms before exporting. |
| Short-form social video | Works well | Short caption mode and word-by-word timing are built for TikTok, Reels and Shorts. |
| AI captions in non-English languages | Not supported yet | The speech model is English only. Use the Video Editor to import an SRT or VTT file in another language. |
| Noisy audio, strong accents, music | Accuracy drops | Review the transcript carefully. A larger Whisper model improves accuracy if your device can handle it. |
| Long recordings (over 30 minutes) | Possible, needs memory | Browser memory is the practical limit. Split very long files into parts. |
| Multiple speakers | No speaker labels | Captions do not separate speakers. You can edit lines manually after generating. |
| Cloud AI features (background removal, tracking) | Not supported | Those need cloud processing. ToolDeck is built for local, private work. |
Tips for more accurate captions
Pick the language manually for short clips
Automatic language detection listens to the first seconds only. If your clip opens with music, choose English explicitly.
Review names and brand words first
Speech recognition slips most on proper nouns. Search the caption list for names before exporting.
Keep captions short for vertical video
One line in the lower third leaves the face and platform buttons clear. Short caption mode does this for you.
Export the SRT as well as the MP4
Burned-in captions cannot be edited later. Keep the SRT or VTT file so you can re-upload or re-edit captions anytime.
ToolDeck vs other auto caption tools
Cloud caption tools upload your video and render on their servers. ToolDeck trades cloud features for privacy and no limits: nothing is uploaded and exports have no watermark. Choose by what your project needs.
| Feature | ToolDeck | Typical cloud caption tools |
|---|---|---|
| Your video is uploaded to a server | Never | Usually yes |
| Watermark on captioned video | None | Often on free plans |
| Sign-up or account required | Not required | Usually required |
| Free plan video length cap | None from ToolDeck | Common: 3–10 min |
| Word-level caption timing | Yes | Varies by tool |
| Export SRT and VTT subtitle files | Yes | Varies by plan |
| Burn captions into MP4 | Yes | Often paid |
| Animated caption styles | 40+ styles, on device | Yes |
| Works offline after loading | Yes after model download | Usually no |
| AI caption language | English speech today | Often multilingual |
| Cloud templates and stock assets | No | Yes |
General comparison. Individual products differ, and features, limits and pricing change.
Caption content you have the right to use
Only caption and publish audio you own or are licensed to use. Captions reproduce spoken content, so make sure you have permission to transcribe and publish the speech in the video. You are responsible for how you use the result.
More free video tools
Caption, trim, silence-cut and edit. Nothing uploaded, ever.
Auto Caption Generator
You are hereAI captions with SRT and VTT export, no upload.
Video Editor
Multi-track editor with AI captions and filters.
Remove Silence from Video
Cut dead air automatically.
Add Subtitles to Video
Import SRT or VTT and burn subtitles.
SRT Generator
Turn speech into SRT files.
CapCut Alternative
Free browser editor, no watermark.
Your video and audio never leave your device
There is no upload step and no account. Transcription and export happen in your browser. The open speech model is downloaded once from a public CDN and runs on your device. You can check this yourself: open your browser developer tools, go to the Network tab, then generate captions and export a video and watch for file uploads. Read how we keep it private.